The research will explore urine control problems in people with some type of dementia living at home. No work has been done to document effective nursing care strategies in caring for dementia patients at home. The research will show that urine control problems in dementia patients are most frequently related to functional difficulties. Further, it will demonstrate the efficacy of helping caregivers to compensate for functional impairments when providing continence care by implementing individualized prompted toileting. The usefulness of two methods of implementing the program with caregivers will be evaluated to determine the immediate and long term effectiveness of the methods. The long term objective is to identify nursing strategies that will lessen the burden on caregivers of dementia patients and prolong the time caregivers are able to maintain dementia patients at home. Findings from the project can potentially improve the quality of life of dementia patients and their caregivers. Subjects will consist of incontinent dementia patients and their caregivers living at home who respond to a newspaper announcement calling for volunteers. Subjects will be randomly assigned to one of three groups - intervention/visit, intervention only, and control. Subjects in the two intervention groups will be taught how to implement individualized prompted toileting to decrease the incidence of urinary incontinence at home. Subjects in the intervention visit group will also be visited bimonthly by a professional nurse. The intervention/visit group will be aided by the nurse in implementing the program at each visit. Subjects in the control group will receive no intervention. All groups will be followed for a 30- month period. At the beginning of the study the following data will be collected: demographic data, physical examination findings, urinanalysis, mental status, mobility measures, activities of daily living, functional abilities related to urine control, perceived caregiver burden and incidence of urinary incontinence. Growth curve analysis will be used to describe changes in persons with dementia over a 30-month period. Multiple regression equations will be constructed predicting the growth curve coefficient from treatment group, treatment implementation, caregiver burden, and patient variables.